How AI is reworking the solar power business? – NEWS, POWER

How AI is transforming the solar energy industry? - NEWS, POWER

With the rapid development of ICT technologies like AI, cloud, big data and 5G, Huawei has partnered with experts in the field and published 10 new technical trends for solar energy and smart PV for 2025.

Full intelligence: In the future, data will become the most basic resource for the digital world, and all data in the PV system from modules, trackers and inverters should be recorded. We assume that 90% of PV systems will be completely digitalized by 2025.

We expect AI to be used in optimizing the system design for detection, collaboration in storing solar storage, and ultimately improving the intelligence of the entire system. By 2025, 70% of PV systems will introduce AI.

We believe that unmanned operation and maintenance will be the mainstream. Drones are used for the inspection, and auxiliary robots can help repair and replace remote control devices. We assume that by 2025 80% of the PV systems will be unmanned, at least with light manual operation.

Grid support: Currently, renewable energy installation capacity accounts for 10% of total power plants, with power generation currently accounting for only 2%. In the future, renewable energy will become the main source of energy as significantly more PV systems are connected to the power grid.

The storage of the PV system is expected to be a new trend for grid support. For 2025, we expect that 30% of the new PV systems will have an energy storage facility. However, the intelligent collaboration between solar and storage to improve system efficiency will be another new project for PV solution providers.

In addition, 80% of residential systems will connect to VPP networks (Virtual Power Plant) by 2025.

Update system and components: With the development of the material and heat dissipation and algorithm, we expect the components in the PV solution to be upgraded in three ways. Modular structure of components for flexible provision. Second is the power density. In the future, the power density of the inverter will increase by 50%. Ultimately, reliability, availability, security, resilience and data protection become essential requirements.

Based on these findings, Huawei has developed the FusionSolar 6.0+ Utility Smart PV solution, which consists of intelligent inverters, a controller for data transmission and control, the Smart Transformer Station for the IPS solution overseas and a Smart PV management system for the Real-time monitoring and management exists. Our latest enhancements to the solution are examples of how we are using AI technology in FusionSolar: 1) Smart Tracker Control Algorithm, 2) Smart IV Curve Diagnostics 4.0, and 3) Smart Grid Connection Algorithm.

Smart tracker control algorithm:

The tracker can significantly increase energy production. More and more PV systems are using trackers to reduce the LCOE. Currently, the trackers rely on astronomical algorithms to ensure that the modules are perpendicular to direct sunlight. However, the uneven landscape in the PV system is not taken into account. The shading between different rows of PV modules could reduce the yields. On cloudy days the diffuse light is higher than on normal days. And the tracking angle without optimization cannot bring in as much power when the light is scattered.

Huawei integrates AI and collaborative power generation into a tracking algorithm to optimize the tracking angle through the repeating neural network based on the bulk data of the plant performance. And during the backtracking, Huawei takes over each row to avoid shadows between the rows and to gain more solar energy.

The results show that the Smart Control algorithm can improve energy production by at least 1% compared to the standard mode. This technology uses standard interfaces and communication protocols and it is quite easy to work with different tracker vendors.

Intelligent IV curve diagnosis

We are also using AI to reduce OPEX as we improve IV curve diagnosis. The intelligent IV curve diagnosis can significantly reduce the inspection time on site from weeks to half a day. AI learns based on expert knowledge and can precisely identify 14 types of string failure certified by TÜV. These diagnostics can also provide a troubleshooting solution to close the O&M process. This feature has already been used in a 7 GW PV installation around the world and the number continues to grow.

Huawei continues to update the algorithm to accommodate more and more scenarios. This year the 4th generation can support the diagnosis of the mono, bifacial and shingle PV modules. It is also compatible with the mixed installation of different PV modules in one string.

Smart grid connection algorithm

With the LCOE of PV continuously decreasing, solar is likely to be the most important source of energy, according to forecasts by the International Energy Agency. In the near future, over 30% of the new power plant installed each year will be the PV system. However, with the increase in the build-up of renewable energy, the short circuit ratio (SCR) of the power grid is gradually decreasing. We could see that the SCR will decrease significantly in some countries. Or in another way, more and more regions will face a new challenge from a weak network. The SCR is expected to be below 2 in many countries between 2025 and 2030.

Huawei has been researching network connection technologies for many years and improving their control accuracy. Huawei has currently achieved a stable grid connection in all PV scenarios. We are confident that our algorithm can help our customers to build a robust system with higher reliability. The smart inverters could support SCR of only 1.5 to avoid common synchronization errors. In addition, the Huawei solution offers better POD (Power Oscillation Damping) control, which some of our customers have already considered and partially adopted.


Please enter your comment!
Please enter your name here